Federal, state, and local governments as well as private industry are planning and implementing widespread drug abuse testing and treatment programs. However, there continues to be a general lack of empirical evidence to help focus these plans for reducing drug abuse. Therefore, the proposed research project is intended to increase our knowledge base concerning drug use in the workplace. First, it will estimate drug and alcohol use prevalence among a representative sample of municipal employees in Houston. These rates will be studied in terms of sociodemographic and job classification characteristics. Second, it will identify work and non-work factors associated with substance use, along with employee attitudes and perceptions of the effects it has on work performance. The project will be conducted in collaboration with the City of Houston, which currently employs a total of about 21,000 workers. For this study, a sample of 2,500 employees will be selected using a stratified sampling plan to represent departments and job types. Management, skilled/technical, clerical, and skilled job classifications will be included. Self-report questionnaires will be administered anonymously to intact employee workgroups by members of the research team; information will include sociodemographic characteristics, patterns of personal use of alcohol and other drugs (including while on-the-job), perceptions of drug use by co-workers, attitudes about drug use effects and policies, job satisfaction, attributes of the work environment and job performance indicators. Workgroups will also be randomly selected to yield a 25% subsample (N = 625) of employees for obtaining urine specimens anonymously to validate self-reported drug use. Participation in the study will be voluntary (but using incentive mechanisms), and procedure will be carefully explained to participants and followed to protect the confidentiality of data City personnel will not be involved in the data collection or analysis process, and will not have access to any individual employee data files. Information will be reported only in aggregate form but with input and cooperation of management and employee representatives to help address informational needs of these groups. The planned duration of the project is 24 months, with descriptive feedback to city planners (especially the health and insurance division) scheduled to begin early in the second year.